Insect. Soc. 50 (2003) 401–402 0020-1812/03/040401-02 Insectes Sociaux DOI 10.1007/s00040-003-0697-x © Birkhäuser Verlag, Basel, 2003

Short communication

Evidence for mating plugs in the fire Solenopsis invicta

A. S. Mikheyev Section of Integrative Biology, The University of Texas at Austin, 1 University Station C0930, Austin, TX 78712-0253, USA, e-mail: [email protected]

Received 16 December 2002; revised 11 June 2003; accepted 18 June 2003.

Summary. Male inhibition of female re-mating is com- 2001), impossible. Thus, evidence for the existence of mat- mon in many . Mating plugs, used by males to con- ing plugs can only be indirectly inferred. Specifically, if the trol female re-mating, have been postulated in several ant MAGs are homologous to those of bumblebees and serve ant . Recent studies of bumblebees have described the same function, one would expect their contents to be re-mating inhibition by male accessory gland secretions. chemically similar, i.e. to contain fatty acids. Furthermore, it Fire Solenopsis invicta possess accessory glands con- should be possible to detect the transfer of the secretion into taining the same four fatty acids as do the bumblebees. Fur- the female after mating. thermore it appears that some of these acids are transferred to the female at mating. Thus, it is possible that single mat- ing of fire ant females may be enforced by male mating Materials and methods plugs. In order to determine the composition of fire ant secretions, MAGs were dissected, ruptured into hexane and analyzed by GC/MS, following Key words:Fire ants, male accessory glands, mating plugs. methylation with methanolic hydrochloride. The methylation was also carried out on synthetic stearic acid to determine approximate yield for the reaction. All data presented are extrapolated to 100% theoretical yield based on this synthesis. In order to determine transfer of MAG Introduction secretions into the female, the internal reproductive organs (i.e. the bur- sa copulatrix and spermatheca) of 20 newly mated (1–3 h) and 20 virgin females were dissected and analyzed by GC/MS as above. The Single insemination of queens appears to be the norm relative amount of secretion in mated and unmated females was ana- in many hymenopteran social insects, despite numerous lyzed by ANCOVA using queen live weight as a covariate. advantages postulated for polyandrous females (Strass- mann, 2001). The lack of polyandry can be explained adap- tively either by the absence of selection favoring multiple Results mating by females, or by selection on males to oppose polyandry. The latter appears to be the case in the bumble- The S. invicta MAGs contained a mixture of fatty acids bee Bombus terrestris, where polyandrous females have (linoleic, oleic, palmitic and stearic acids), as well as small higher fitness, but are typically prevented from re-mating by amounts of benzoic acid. Taken together, these fatty acids male ‘mating plugs’, which consist of a lipid cocktail secret- make up around 20% of the total dry MAG weight. The live ed by male accessory glands (MAGs) (Duvoisin et al., 1999; weight of the queens did not correlate with the amount of fat- Baer et al., 2000). The existence of mating plugs has also ty acids contained inside their copulatory organs (P > 0.20). been speculated for several ant species (see Bourke and The amount of benzoic acid was not different in the mated Franks (1995) for a review). The fire ants Solenopsis invicta and unmated queens (P > 0.15). Although the mean amount possess MAGs that resemble those of bumblebees (Ball and of palmitic acid was higher in mated queens, this effect was Vinson, 1984), which raises the possibility that their single suggestive but not significant (0.07 < P < 0.08). However, mating (Ross and Fletcher, 1985) may also be mating plug mated queens contained much greater amounts of linoleic enforced. (P < 0.001), oleic (P < 0.001) and stearic (P < 0.003) acids. Fire ant mating takes place high in the air and is difficult The copulatory organs of newly mated queens contain about to observe, rendering manipulative mate choice studies, such 16 ng of the latter three fatty acids, much more than those of as those performed on the bumblebees (e.g., Sauter et al., unmated queens (3 ng). The amount of secretion transferred 402 A.S. Mikheyev Fire ant mating plugs

done on the mechanism of re-mating suppression via lipids and the possibility of female counter-measures. Based on structure, position and the content of their secretions, it appears that the MAGs of bumblebees and the fire ants are homologous. If mating plugs are found in both ants and bumblebees, MAG-mediated queen re-mating sup- pression may be widespread and ancient in the hymenopter- an social insects and may have played an important role in the evolution of complex relatedness-based social systems.

Acknowledgements

I am grateful to W.R. Tschinkel and his lab for their support throughout the project and assistance with laboratory equipment and supplies. I thank A.F.G. Bourke and an anonymous reviewer for their comments on an earlier version of the manuscript. I am indebted to R. Rogers for allowing me to use the GC/MS at the National High Magnetic Field Figure 1. Seminal vesicles (SV) and male accessory glands (AG) of S. Laboratory and C.A. Hughey for technical assistance with the device. invicta Funding for this project has been provided by an Explorer’s Club grant and a Sigma Xi Grant-in-Aid of Research. to the females is on the same order as the total amount of lipids estimated in S. invicta MAGs (20 ng). References

Baer, B., R. Maile, P. Schmid-Hempel, E.D. Morgan and G.R. Jones, Discussion 2000. Chemistry of a mating plug in bumblebees. J. Chem. Ecol. 26: 1869–1875. The contents of fire ant MAGs consists of the same three or Baer, B., E.D. Morgan and P. Schmid-Hempel, 2001. A nonspecific fat- four fatty acids that make up more than 90% of the bumble- ty acid within the bumblebee mating plug prevents females from bee secretion’s weight (Baer et al., 2000). In particular, the remating. Proc. Natl. Acad. Sci. USA 98: 3926–3928. Baer, B. and P. Schmid-Hempel, 2000. The artificial insemination of males transfer linoleic acid, which was experimentally bumblebee queens. . Soc. 47: 183–187. demonstrated to inhibit re-mating by bumblebee queens Ball, D.E., J.T. Mirenda, A.A. Sorensen and S.B. Vinson, 1983. Instru- (Baer et al., 2001). Though the present study cannot defini- mental insemination of the fire ant Solenopsis invicta. Ent. Exp. tively rule out other male organs as sources of the fatty acids Appl. 33: 195–202. (e.g. the seminal vesicles), a study of several carpenter ant Ball, D.E. and S.B. Vinson, 1984. Anatomy and histology of the male reproductive system of the fire ant, Solenopsis invicta Buren species has found accessory glands to contain major stores of (: Formicidae). Int. J. Insect Morphol. Embryol. 13: carbohydrates (Wheeler and Krutzsch, 1992), suggesting 283–294. that the lipid secretions are indeed MAG-derived. The obser- Bourke, A.F.G. and N.R. Franks, 1995. Social Evolution in Ants. Prince- vation that fire ant MAGs contain and transfer the same fat- ton University Press, Princeton, New Jersey. 529 pp. ty acids as do those of the bumblebees makes it likely that Brown, M.J.F., B. Baer, R. Schmid-Hempel and P. Schmid-Hempel, they serve they both serve to inhibit female re-mating. Fur- 2002. Dynamics of multiple mating in the bumblebee Bombus hyp- norum. Insect. Soc. 49: 315–319 ther circumstantial evidence linking the function of fire ant Duvoisin, N., B. Baer and P. Schmid-Hempel, 1999. Sperm transfer and and bumblebee MAGs comes from reports that MAG secre- male competition in a bumblebee. Anim. Behav. 58: 743–749. tions are not critical for the successful instrumental insemi- Robertson, H.G., 1995. Sperm transfer in the ant vidua Smith, nation of either bumblebees (Baer and Schmid-Hempel, F. (Hymenoptera, Formicidae). Insect. Soc. 42: 411–418 2000) or fire ants (Ball et al., 1983). Ross, K.G. and D.J.C. Fletcher, 1985. Comparative study of genetic and social structure in two forms of the fire ant, Solenopsis invicta. The mere presence of a mating plug may not be enough Behav. Ecol. Sociobiol. 17: 349–356. to guarantee single insemination. For example, mating plugs Sauter, A., M.J.F. Brown, B. Baer and P. Schmid-Hempel, 2001. Males are present in the naturally polyandrous bumblebee B. hyp- of social insects can prevent queens from multiple mating. Proc. R. norum (Brown et al., 2002). Similarly, if mating plugs dimin- Soc. Lond. B Biol. Sci. 268: 1449–1454. ish re-mating in ants, they may not be completely successful Strassmann, J., 2001. The rarity of multiple mating by females of social Hymenoptera. Insect. Soc. 48: 1–13. in all cases. For instance, male secretions that block the vagi- Wheeler D.E. and P.H. Krutzsch, 1992. Internal reproductive system in nal orifice in the ant Carebara vidua do not prevent multiple adult males of the genus Camponotus (Hymenoptera: Formicidae: mating (Robertson, 1995). Thus, much more work must be Formicinae). J. Morphol. 211: 307–317.